Last updated:
Gemini 4 Argon vs Claude Opus 5.5
Google's new frontier model against Anthropic's top general-release model. At standard prices they cost exactly the same per token.
Side by side
| Gemini 4 Argon | Opus 5.5 | |
|---|---|---|
| Vendor | Google DeepMind | Anthropic |
| Released | September 30, 2026 (Fairwind Program only) | September 22, 2026 |
| Input / output per 1M | $2 / $10 launch, $4 / $20 standard | $4 / $20 |
| Cached input per 1M | $0.10 launch, $0.20 standard | $0.20 |
| Max output | 1M tokens | 128K tokens |
| Context window | — | 1M |
Benchmarks
From Google's Gemini 4 Argon table. Google ran every model at its highest thinking setting. Anthropic publishes its own comparisons, which can differ.
| Benchmark | Gemini 4 Argon | Opus 5.5 |
|---|---|---|
| Vals Index | 68.9% | 67.0% |
| AutomationBench | 51.3% | 42.5% |
| Vals Finance Agent v2 | 65.4% | 58.6% |
| Harvey's Legal Agent Benchmark | 19.6% | 3.8% |
| DeepSWE v1.1 | 77.9% | 74.2% |
| FrontierSWE v2 | 55.0% | 62.3% |
| Vibe Code Bench | 91.9% | 90.3% |
| Terminal-Bench 4.0 | 57.4% | 66.4% |
| PostTrainBench | 45.3% | 49.3% |
| Terminal-Bench Science 0.1 | 57.6% | 63.3% |
| LABBench 2 | 88.8% | 73.1% |
| RiemannBench | 76.0% | 69.6% |
| GraphWalks BFS (up to 128K) | 99.7% | 90.6% |
| GraphWalks BFS (256K-1M) | 84.2% | 66.8% |
| Agent's Last Exam | 39.5% | 38.2% |
| OSWorld-2.0 | 69.2% | — |
| Chartography | 71.6% | 66.3% |
| LVBench | 91.7% | 83.7% |
| CWE-bench v1 | 68.0% | 67.0% |
Argon leads on 14, Opus 5.5 on 4, and 0 are ties.
Source: Google DeepMind: Gemini model page
Cost per coding task
| Codebase size | Argon (launch) | Argon (standard) | Opus 5.5 |
|---|---|---|---|
| Small (< 10k lines) | $0.112 | $0.223 | $0.223 |
| Medium (10k-100k lines) | $0.249 | $0.498 | $0.498 |
| Large / monorepo (> 100k lines) | $0.594 | $1.188 | $1.188 |
At 173 tasks a month on a medium codebase: about $43 on Argon at the launch price ($86 at the standard price) and $86 on Opus 5.5, with the same token counts for both.
Gemini 4 Argon: full details Compare every model in the calculator
Which one to pick
- Terminal-heavy coding agents today: Opus 5.5, which leads on Terminal-Bench 4.0 and works in Claude Code now.
- Legal, finance and business automation: Argon scored far higher on Harvey's legal benchmark and AutomationBench (51.3% vs 42.5%).
- Very long inputs: Argon, on GraphWalks from 256K to 1M (84.2% vs 66.8%).
- Budget: Argon's launch price is half of Opus 5.5, once you can get it.
FAQ
Is Gemini 4 Argon better than Opus 5.5?
On Google's table, Argon beats Opus 5.5 on 14 of 18 shared benchmarks. Opus 5.5 is ahead on FrontierSWE v2, Terminal-Bench 4.0, PostTrainBench, Terminal-Bench Science 0.1. Vendors pick benchmarks that suit them, so test on your own work.
Which is cheaper, Gemini 4 Argon or Opus 5.5?
Argon at its launch price ($2 / $10). At its standard price ($4 / $20) it costs the same as Opus 5.5.
Can I use Gemini 4 Argon instead of Opus 5.5 today?
Not unless you are in Google's Fairwind security program. Paid Gemini API customers and Google AI Ultra subscribers are next, with no date given.
Independent page, not affiliated with Google, Anthropic or Meta. Benchmarks are Google-published; we have not run our own tests.